I feel like we are moving into the robocop plot very quickly haha
ETHICS
-
LLM Evaluators: Blind Spots and Adversarial Vulnerabilities Exposed
By
–
References: – Finding Blind Spots in Evaluator LLMs with Interpretable Checklists https://
arxiv.org/abs/2406.13439
v1
…
– Is LLM-as-a-Judge Robust? Investigating Universal Adversarial Attacks on Zero-shot LLM Assessment https://
arxiv.org/abs/2402.14016
– On the Limitations of Fine-tuned Judge Models -
ARMANN adapts human-centered AI solutions for real-world applications
By
–
Ms. Amrita Mahale, Head of Product at ARMANN, discussed how they are adapting their AI solutions at ARMANN to be human-centered and user-focused in the 'IndiaAI: Real World AI Solutions' session @abhish18 @GPAI_PMIA @_DigitalIndia @GoI_MeitY #ai #aisummit #aijourney
-
Building Bias-Free LLMs for Diverse Languages and Cultures
By
–
In the session, “ IndiaAI: Large Language Models” at #GlobalIndiaAISummit 2024, Dr. Kalika Bali discussed the challenges of building bias-free LLMs for diverse languages and cultures. @kalikabali @abhish18 @_DigitalIndia @GPAI_PMIA @GoI_MeitY #ai #aisummit #aitechnology
-
Humanity’s Innovation: From Haber Process to AI Salvation
By
–
Humans are the species that keeps growing because it always finds an innovation to save the day when it's on the brink of catastrophe. In the 20th century it was the Haber process. In the 21st it will be AI.
-
AI arms race threatens BigTech net-zero climate pledges
By
–
Great article on how the current AI arms race is making is increasingly unlikely that BigTech will meet their net-zero pledges. Arms races made humanity smart through evolution, but will it also be its demise? Source: the Guardian https://
search.app/szor4fvupc85zT
g99
… -
AI Emissions Rising Challenge Company Climate Goals
By
–
The whole report is really interesting; suggest taking a close look at pages 12 and 31 for some more specifics related to how AI contributed to emissions rising and is expected to make it tricky to meet company goals going forward.
-
AI Moratorium Would Have Been Necessary, Now Too Late
By
–
I admit it, I was wrong – only an AI moratorium could have saved us. And now it's too late.
-
AI Ethics Committees Lack Biodiversity Impact Measurement Standards
By
–
AI Ethics committees embracing LLMs where no accurate measurement of biodiversity usage are indicated = BS.
-
Tech Companies Environmental Impact Reporting and Transparency Gaps
By
–
i appreciate these reports for 2 reasons: it gives a sense of what's going on at these companies (enviro impacts of AI have in the past been tricky to track) and it makes me think about alllllll the stuff the *other* tech companies are doing that we don't know anything about.